Sorry Lou but life's a Grich, amirite?

While Evers’s numbers are anything but great, it should be noted that he won an MVP award and played for three World Series-winning teams (the Cubs twice and the Boston Braves once). That said, it’s highly unlikely that he would be in the Hall of Fame were he not part of a legendary double-play combination

Evers isn't in my PHoM, but he's 27th at 2B all time, which is in the borderline. And he was at worst 7th best at his position when he retired. Tinker is solidly in my PHoM, and Chance is also a borderline guy but higher than Evers.

Sorry Lou but life's a Grich, amirite?

I would love to see Lou in the Hall, but Grich is a better candidate.

if you look at the Miracle Braves of Boston in 1914, and the prior track records of the main players, and their performance that year, and the arrival of Evers that same season......

I struggle to just ignore that.

(btw, it amuses me to no end how much Evers and Tinker hated each other.)

Not too late for some belated Whitaker/Trammel poetry that would impress the Veterans Committee.

Hall of Aim voters are on the hornsby of a dilemma Al, be wise as a fox and induct Whitaker, or just keep sandberging the issue till some frisch faces come along to carew Lou through the doerr of the Hall. And oh what lajoie that will be!

Not too late for some belated Whitaker/Trammel poetry that would impress the Veterans Committee.

Alan to Lou to Jason/Richie/Enos/Dave/Darrell/Cecil doens't quite have the ring to it.

Wow, Jason Thompson was all the way back in the mid-late 70s? I would have put him at least a decade later.

Alan to Lou to Jason/Richie/Enos/Dave/Darrell/Cecil doens't quite have the ring to it.

We don’t have to limit ourselves to emulating Franklin Pierce Adams; William Blake should be good enough:

Tigers, Tigers burning bright
In the ballpark day & night,
Whitaker & Trammel, the fans delight,
Greatest keystone combo does all excite.

One enshrined, the other slighted
That wrong to be righted,
And sabermetricians delighted,
Let Lou be Cooperstown knighted.

The Nobel Prize folks (and others) may not be impressed, but good enough for the likes of the Hall of Fame Contemporary Era Committee.

Admittedly, the committee was won over by

Sphan not Sain
but (surprise!) Baines

if you look at the Miracle Braves of Boston in 1914, and the prior track records of the main players, and their performance that year, and the arrival of Evers that same season......

I struggle to just ignore that.

I'm glad you mentioned this. I'd heard it mentioned before in Tinker's favor. He also performed really well in both WS wins for the Cubs and the WS for the Braves.

#7 - how about Alan to Lou to ... who?

The dilemma the Tigers face.
Alan to Lou to who?
First was never secure, but short and second were.
Alan to Lou to who?

Nah, some sort of “who’s on first” thing would never catch on.

It's always a bad idea to compare your Candidate of Choice for the HoF to one of the weaker members already enshrined. Evers isn't really a bad choice, but he's Outer Circle, without question. As for the impact on the 1914 Braves, that would be more serious if it wasn't clear that the Bg Thing that put the Braves into overdrive was the fluke season by pitcher Bill James, who appears to have blown his arm out in 1914, but was huge in that one year. Evers did combine with superglove Rabbit Maranville to form the most effective keystone duo in the league, but that wasn't as important then as it would be now.

quick summary of 1914 Miracle Braves and Evers' impact:
--
Braves improved by 25 games in 1914. Some of that was clutch/luck.

Evers came in from the Cubs. The Cubs won 10 fewer games in 14 compared to 13.

Evers replaced a 2B named Sweeney. Evers led the legaue in FPCT with 17 errors. Sweeney had made 45.

The Braves imroved immensely in run prevention. Not because their ERA was much better, but because the UNearned runs went from near-worst to league-best. They had mostly the same pitching staff both years.

When you combined the 1914 season, were Evers was named MVP (for good or bad reasons) with Evers role in the 1906-08 champs, you can see why his defense rep is even better than the ##s, lending some credence to his HoF case.


-- Yes, Bill James is a candidate for biggest one-season wonder career in MLB history. 26-7 in 1914. 11-14 otherwise.

Evers replaced a 2B named Sweeney. Evers led the legaue in FPCT with 17 errors. Sweeney had made 45.

I think looking at how he may have impacted SS defense is interesting as well. Maranville was very young, so maybe it was not anything related to Evers, but he did go from 0 runs above average in 2013 to 26 in 2014. Maranville was 9 above average the next year, then 15 the following year, and then down to 5, -2, 5, and 2 in the four seasons after Evers left the team. He also shot back up to well above average after moving to Pittsburgh and playing with a hodgepodge of 2B there though, so not sure how meaningful 1914 may have been in terms of Evers helping Maranville improve.

Wasn't 2B defense more important in that era than in subsequent ones?

Evers did combine with superglove Rabbit Maranville to form the most effective keystone duo in the league, but that wasn't as important then as it would be now.

???? There are fewer balls in play now, so I would think a great keystone combo was much more important in 1914. One in 10 PA ended in a strikeout in 1914, compared to about 1 in 4 now. Flyballs are probably more prominent now but I don't have that data.

Wasn't 2B defense more important in that era than in subsequent ones?

my impression is that in that dead ball era 3b was supposed to be more defensive in nature and 2b had more room for being a "bat first" position. Perhaps that is fueled by Hornsby arriving in 1915 (Lajoie was 39 in 1914). Also of note guys like Heinie Zimmerman, and Heinie Groh were "bat first" guys who played both 2b and 3b in this era (perhaps Buck Herzog might fit that profile as well)

I think the main reason is that in the dead ball era there was a lot more emphasis on staying out of the DP and bunting both to move up runners and for base hits. That put more emphasis on defensive 3b. But that's my guess and Im not sure anyone's come right out and said that. Although its been pointed out that these positions (2b and 3b) do seem to switch places on the defensive spectrum about the time of the arrival of the lively ball.

Going from memory but as I recall this early dead ball era lasted only about a decade so arguably less than the modern "TTO" era that we live in today. And given that fairly brief period perhaps teams really didnt have enuf time to develop talent that would be custom made to fit the roles of 3b and 2b. For example it took at least 5 or 6 years for teams to find power hitters roughly of the same caliber as Ruth. It does take some years for the league to catch up to a changing environment.

EDIT the period from 1904-1919 featured most seasons with runs/game under 4 hence the dead ball era generally. But runs/game soared in 1911-1912 to modern day level of 4.5 So basically you have 7 seasons before that and 7 seasons after that of true dead ball offense. Modern day Ks have been going up since 2006, HRs jumped in 1994 and have remained high ever since. Significant shifting has been with us for what ten years? So these modern trends are arguably longer lived than the relative short dead ball intervals.

???? There are fewer balls in play now, so I would think a great keystone combo was much more important in 1914.

I agree with Barry here. Im struggling to understand what BJ meant in post 13 when said the defense wasnt as important. Were there less double plays in those days? still it was important. Sac hits starting dropping in the 20s before bottoming out in 1930 to about half of the dead ball era rate. Presumably there were less DPs in the dead ball era but it looks like the Project Scoresheet people havent gotten around to filling out that part of the statistical record. Also base stealing was at record levels so there must be some need for mid infielders can catch and tag the runner.

From 1905-1909 -- the real dead ball era -- NL 2b turned between 348 and 429 DP/year. In 1910 that jumped to 555 DP/year for reasons I don't know, but the highest for the decade through 1919 was 569 and the lowest was 475 DP/year. Starting in 1920, DP/year jumped to 615 in 1920 and reached 877 by 1930.

@18: It's a pet peeve of mine that people say that 2b was more important than 3b prior to the 1920s (I understand that you're just stating common wisdom). The numbers just don't support this. Even before 1920, 2b made 1000 plays/year more than 3b. That number continued to increase in the 20s, when it was never less than 1500 and reached 2000. This was not because 3b suddenly started making fewer plays -- their numbers stayed remarkably constant from 1905-30. It was that 2b remained constant through 1919 and then started making many more plays.

Folks here have argued that 3b had to make more difficult plays, but even if I grant that (and I'm not sure I should) I don't see how it makes up for that wide a gap in the sheer number of plays made.

That said, there's an argument the managers in the dead ball era *thought* that 3b was more important, but managers have made lots of mistakes throughout the history of the game; this, I think, was one.

@18: It's a pet peeve of mine that people say that 2b was more important than 3b prior to the 1920s ....

Im not offended or anything but where/how did I say that? I was looking at it from the stand point of "Defense first" vs "offense first" type of situation. I would probably never say any single position in the field is "more important." Why would anyone say that?

That said, there's an argument the managers in the dead ball era *thought* that 3b was more important,

again what does that mean? Both positions are presumably equally important. No? Im confused by what you're trying to say here.

Did you mean "more important defensively?"

I guess my next question would be: Given there was less bunting in the 1920s than previous seasons, would 3b evolve over time to be more robust in stature and less need for quickness?

Mefisto: I welcome the discussion on a subject Im sure you're more well versed than. To me knowledge the only thing Ive said re: "conventional wisdom" would be this:

Although its been pointed out that these positions (2b and 3b) do seem to switch places on the defensive spectrum about the time of the arrival of the lively ball.

So just to be clear: is the above statement true or not? did these positions swap places on the so called spectrum?

In my view, the two positions did *not* switch places on the defensive spectrum. 2b was always more important defensively.

Im not offended or anything but where/how did I say that? I was looking at it from the stand point of "Defense first" vs "offense first" type of situation. I would probably never say any single position in the field is "more important."

More important defensively.

Given there was less bunting in the 1920s than previous seasons, would 3b evolve over time to be more robust in stature and less need for quickness?

That's a good question but I don't really know. Home Run Baker was 5'11" 173, which seems pretty good size for that era, and Jimmy Collins was 5'9" 178, but without a systematic data base it's kind of hard to give a definitive answer.

In my view, the two positions did *not* switch places on the defensive spectrum. 2b was always more important defensively.

He was probably somewhat confused by your first line in Post 21.

Ive read this sentence 150 times and Im on the verge of having a stroke:

@18: It's a pet peeve of mine that people say that 2b was more important than 3b prior to the 1920s (I understand that you're just stating common wisdom). The numbers just don't support this. Even before 1920, 2b made 1000 plays/year more than 3b

I dont get it. Can you rephrase that?

In my view, the two positions did *not* switch places on the defensive spectrum. 2b was always more important defensively.

Oh OK. Now we've got a good starting pt. for discussion if there is any to be had.

He was probably somewhat confused by your first line in Post 21.

I thought he was maybe responding to Howie in no. 16.

@25: That's fair.

@26: I mean that the importance of a defensive position is roughly related to the number of plays that position has to make. Specifically, I'm referring to assists because PO data are too contaminated with popups and easy steps on the bag to be useful. So what the data show is that league wide, 2b were making 1000+ more assists every year than 3b in the NL before 1920. Given my assumption that the number of assists has a direct correlation to the defensive importance of the position, that means 2b was more important defensively than 3b; managers should have been putting their better fielders at 2b.

Now, have I *proven* this? Obviously not, though I think my assumption is reasonable. As far as I know, nobody has ever actually shown that 3b is more important defensively. They just assert it (perhaps because Bill James asserted it long ago, though without any data).

I should clarify that I'm talking about the IF positions here. Different standards apply to C and OF.

My question was based on whether Evers excelling at 2B relative to his peers (if he did) would separate him from the pack more than a 3B could do.

I think someone may may made this case in our discussion of Evers what, 15 years ago? hence my question because I am not confident I remember this accurately.

I think 29 is a nice summary and hard to dispute. The only contrary argument I can come up with would be something like: if we think 2b was more preeminent defensive position during this time then shouldn't we see more better hitting emerge from the 3b position. Why no Eddie Mathews instead Pie Traynor?

But theres probly a number of reasons for that. The dead ball period was short and baseball was in a transitional period. Also there as economic pressure from wwi as well as the federal league so perhaps there wasn't enuf time and money for that position to develop the way it has

What about Home Run Baker? He seems to have been badly affected by missing two full seasons, 1915 and 1920, but until 1914 he looks pretty Matthews-ish to me.

From 1905-1909 -- the real dead ball era -- NL 2b turned between 348 and 429 DP/year. In 1910 that jumped to 555 DP/year for reasons I don't know, but the highest for the decade through 1919 was 569 and the lowest was 475 DP/year.

At least part of the increase presumably comes from a higher scoring context. NL OBP was between .306 and .310 from 1906-09; over the next few years it went .328, .335, .340 before cooling off a bit as the spitball was popularized. If memory serves, the ball was changed in 1910 which resulted in a brief surge in offense that didn't stick until the spitball ban and the post-Ray Chapman move to use new balls more often throughout the game.

What about Home Run Baker? He seems to have been badly affected by missing two full seasons, 1915 and 1920, but until 1914 he looks pretty Matthews-ish to me.

I guess but what about after 1920? We dont see many power hitters from that position until much later. Maybe I was confusing things when I mentioned WWI and the federal league.

That 1914 Braves season was truly miraculous. Nothing else comes close. I have tried to do computer replays and can't match the stranger than fiction wackiness of this team!

There were 3 distinctly different phases, each one more unbelievable than the previous:

1) They started the year 4-18. No other team that started a season even remotely close to that poorly went on to win a title, in fact, ever even made it to the postseason. You start out that badly and your season is effectively over.

2) Part 2 - then they suddenly became this mediocre .500 team that went 22-22 over the next stretch, bringing them to July 4th with an overall record of 26-40, still in last place, 15 games back. This in itself is basically unprecedented, teams that start that badly don't usually improve that much within the same season. Note that the 4th of July is usually about the half way point of the season, but at this point the Braves were well short of 77 games. That means lots of double headers to come! BTW, Bill James had a 7-6 record and was allowing a little over 3 runs/game at this break.

3) Then they amazingly went 68-19 the rest of the way, winning almost 80% of their games over more than half a season. James himself went 19-1 over that stretch. And they followed that up by sweeping the heavily favored A's in the World Series.

Miracle season indeed!

The dilemma the Tigers face.
Alan to Lou to who?

Cecil!

Whitaker and Trammell came up together in 1978. Tigers primary first basemen:

1978-79: Jason Thompson
1980-81: Richie Hebner
1982-83: Enos Cabell
(1980-83 also John Wockenfuss and Rick Leach)
1984-87: Darrell Evans/Dave Bergman
1988: Bergman/husk of Ray Knight
1989: Bergman/husk of Keith Moreland
1990-95: Cecil Fielder

I was four years old and too young to complain, but what rational thought process could explain bringing in a 35-year-old Ray Knight in 1988? The Tigers lost the AL East by one game.

Then they repeated the trick next season with Keith Moreland. Go figure.

@34: That sounds right. I was thinking the offensive jump came in 1911 (I don't know why I thought that; that was what was in my brain). Should have looked it up.

whether defense at 3B was truly more important than 2B before approx 1925 or not, it is very easy to see that at least most MLB clubs *thought* it was. And so a third baseman who hit well back then provided more value to his team relative to today, since the position average (or replacement player, whichever you wish to use) was less able. Uber-systems like WAR and win shares are not wrong to reflect this conceptual truth.

I've never liked the idea of adjusting value according to the *offensive* performance at a position. To me, the correct approach is to give a positional adjustment according to the objectively measured defensive requirements of a position. I understand why the offensive measure got used -- historically it was a proxy for the defensive difference which we couldn't measure. But I still think it's conceptually wrong.

[37] Why not just go with Darrell Evans? He has his own HoF case to make, even if he isn’t the best D. Evans of his era to be excluded from the hall.

If second basemen make more plays, more assists, than third basemen that by itself doesn’t tell me that second is a more important defensive position. The throws from second are a lot easier to make, and that should be factored in somehow as well.

Best test for relative defensive skill is to look at a group of players who played both positions. Preferably in the same year. But if all you have is players moving between seasons, the direction is important. If old second basemen are switching over to third, that implies third is an easier position, etc.

I agree that assists themselves can't prove the issue. We just don't have any good way to measure the relative difficulty of plays made in the period before 1920 (or even after, for that matter). My view is that the sheer number of extra plays being made by 2b (125/team/season) makes it unnecessary to deal with the issue of relative difficulty and we shouldn't let the perfect be the enemy of the good.

I also agree that shifting positions is a good way to handle the issue, in general. It can get tricky, though, if managers have an erroneous belief about relative difficulty and move their worse defensive players to 2b even when they shouldn't.

I suspect much of Sunday Silence's confusion stems from Mefisto's opening typo: @18: It's a pet peeve of mine that people say that 2b was more important than 3b prior to the 1920s ... of course every other word Mefisto wrote argued precisely this. :-) Obviously his pet peeve is people say that 3B was more important ... such as when they say my impression is that in that dead ball era 3b was supposed to be more defensive in nature and 2b had more room for being a "bat first" position.

Proofreading ... it's not just for breakfast anymore! :-)

On play difficulty ... an exceptionally crude proxy (or maybe totally useless) for the difficulty of the plays might be the proportion of RHB. I have no idea if this was much higher in the early days or not but GB to 2B by RHB are not usually hit with much authority. I'm also curious what SS numbers looked like in that era -- was the 3B responsible for more of the SS hole, were teams of that era playing something closer to a shift with the 3B well off the line and the SS more up the middle?

But sure, in the absence of other compelling evidence, the gap in the number of plays made seems a good indicator.

Great summary by Walt there. Yes.

There were certainly a lot less LHB in the dead ball era. The number has been steadily increasing upward probably since before 1920 and didnt level off until perhaps 1950 or later. Im guessing somewhere around 10% in 1910 and about 35% by 1950. Just a guess.

my impression was that the 3b was playing in most of the time because the bunt seems to have been an important tactic in the dead ball era. But Walt's suggestions are also interesting/possible.

...was the 3B responsible for more of the SS hole, were teams of that era playing something closer to a shift with the 3B well off the line and the SS more up the middle?

Going from memory there werent as many LHB in the pre 1920 era. I would have to think the concept of shifting may not have been developed but then again McGraw and Hughie Jennings were doing lots of modern stuff so maybe. I dunno.

Also the success rate for SB drops precipitously in 1931 I noticed that when perusing the offensive trends. God what was that about? I get less base stealing attempts as a normal response to the increase in power but 58% (in '31) what the hell is that about? why bother at all? Success rates dont return to realistic level until 1938, and even that at even lower attempt rate. Attempts dont really get back to modern levels until the late 50s. I guess that's the reason Bill James talked about the 40s and 50s as "station to station" etc.

Look at what a first baseman's mitt looked like in 1900

https://en.wikipedia.org/wiki/Fred_Tenney#/media/File:Tenney.jpg

The jump in the 1920s is significant

https://www.glovedoctor.com/single-post/2017/08/11/how-hank-greenberg-changed-the-baseball-glove

You aren't scooping bad throws from 3B with that 1900s mitt. Accurate throwing was more important then.

but to go from 1900 to 1935 or whatever is 35 years. Do we have a record of how the glove changed over that period of time?

According to Chris Dial, the big improvements in glove technology were in the 1930s. But there were consistent improvements before that as well. Chris may track all of them; I don't know.

According to Chris Dial, the big improvements in glove technology were in the 1930s. But there were consistent improvements before that as well. Chris may track all of them; I don't know.

You can trace the yearly progression of glove technology by simply looking at the ads in the back of the Reach and Spalding guides. I've got a complete 20th century set of both of those publications, and if I weren't so lazy I'd scan them all and give them to someone who'd know what to do with them.

But with just a brief glance at every 10 years from 1910 to 1940, it looks as if Chris has something there, although the biggest breakthroughs didn't come along until the late 1950's, with the Wilson A2000 and the MacGregor GF10 FieldMaster. The GF10 was the first fielder's glove whose web was laced only on the sides, and not at the bottom, which made it much harder for hard hit balls to escape.

Wasn't one of the biggest improvements in glove design the Bill Doak Rawlings glove circa 1920 that introduced the web pocket between the thumb and first finger?

@18: It's a pet peeve of mine that people say that 2b was more important than 3b prior to the 1920s (I understand that you're just stating common wisdom). The numbers just don't support this. Even before 1920, 2b made 1000 plays/year more than 3b.

Hello Mefisto: I wanted to continue this discussion a little more...

I think you were saying that 2b was as important (or more important) defensively as 3b in the dead ball era say 1913-19. Correct me if Im wrong as I think you miswrote something. Anyhow: you seem pretty committed to the idea but there is data that challenge this: Looking at positional adjustment on baseballref in 1916 (and prior years) 2b get no positional adjustment they are right on 0. Then gradually, one run at a time, by 1922 they are +5 runs.

3b in contrast seems to be right at +5 this entire period. I looked at Larry Gardner and Eddie Collins since they seem to have played exclusively at these positions during these years, perhaps the NL has different numbers.

ANyhow. Is that a sort of challenge to what you are saying? Clearly something is changing about the way 2b is being played or something about the way 2b is being viewed.

I mean I appreciate what you are saying, but is this something that should be addressed or explained to make your case?

I'm saying that 2b was more important as a defensive position than 3b at all times from at least 1905 on, and that the importance of 2b increased steadily at least through 1930 (the last year I compiled the numbers). The assist numbers reflect this steadily increasing importance of 2b as we move from 1905-1930. In 1905-1909, 2b averaged 1163 more assists/season than 3b. By 1926-30 that had grown to 1868. In 1922, when BBREF has the positional adjustment equal, NL 2b had 1889 more assists than 3b (236/team/season). I can't see any factor which would make up for having to handle that many more chances. I think both your suggestions are true: something in the nature of the game made 2b increasingly important; and that while it's true that at least some managers in the deadball era seem to have viewed 3b as more important, I don't know of any reason to think that attitude remained by the 1920s.

I don't know why BBREF sets the positional adjustment the way it does for those years; maybe Rally does. If they have data from players who switched positions, that would be good evidence that I'm wrong (not conclusive, but good).

I was thinking the positional adjustment was based simply on how those batters compared to league average. Say for example 3b in the deadball era were maybe 5 runs worse on offense then the average MLB position player.

Therefore it stands to reason that since they hit worse it must be harder to play 3b. This seems like such an obvious way to do this. I mean what else can those numbers mean. But I guess they dont do it like that. They do it as you say they compare players plying multiple positions. Here is Fangraphs def'n:

https://library.fangraphs.com/misc/war/positional-adjustment/#:~:text=In essence, the positional adjustment,than an average first baseman.

but here is baseballref's def'n, and it seems to be a hybrid of what I would do and also compares players who changed positions ("...When one quantifies these differences and also looks at the changes in fielding performance when players move to different positions")

https://www.baseball-reference.com/about/war_explained_position.shtml

Also the baseballref link shows the year by year positional adjustment. You can see in graphic detail the great jump in 2b positional adj. from 1916 to 1922. Also 3b have two drops in this time period. One in 1926-32 and a second drop 1936-43.

Note also that bunting decreased drastically in this period. There is a huge drop in sac hits from 1922-24 about a 25% drop and another drop in 1930. By 1930 its less than half what it was in the 1914-23 period. Presumably there was no longer a great need for a lightening quick 3b. That might explain some of the decrease in 3b adjustment in 1926-32. Probably a lot of other things are going on, that cant be the whole thing.

. The assist numbers reflect this steadily increasing importance of 2b as we move from 1905-1930. In 1905-1909, 2b averaged 1163 more assists/season than 3b.

I think you put too much stock in this method. For instance compare SS to 2b, SS makes perhaps 6-7% more assists. But throughout history SS seems to have positional value about 5 runs more. Thats a big difference and the assist numbers dont reflect that at all. Obviously the difference in assists is greater between 2b and 3b but then you have to deal with the longer throw from 3b and also throws on the DP etc.

The other thing is that if the positional adjustments reflect what I think they do, then it seems 3b were hitting worse than 2b at least in the 1913-22 era. How is that possible if 3b is easier to play? If its easier then those players should be hitting better. Thats true for every position for any time period. I think that has to be addressed to make a solid argument.

To take your last point first, one reason I think that 3b hit worse than 2b in the deadball era is because managers wrongly believed that 3b was more difficult defensively. Wrong tactics appear surprisingly frequently in MLB history -- power hitters shouldn't walk; leadoff hitters should be speedy and we don't worry about OBP; and various opposite attitudes towards SBs. I think that the "3b is more important" belief belongs in this category. Note also that this is a self-reinforcing mindset -- if the manager thinks 3b is more important, he will be more inclined to put a weak hitter there. I also think that using offensive performance to assess defensive difficulty is conceptually wrong even if it's sometimes pragmatically necessary, and this potential for bad strategy by managers is one reason why I think that. All that said, however, the positional switch data would undermine my arguments and a positional adjustment based solely on that would be one I'd accept.

As for SS, there are 2 key differences (at least) when it comes to the assists: SS are just as responsible as 2b for turning the DP, something 3b isn't; and the longer throws require more athletic ability. Those need to be factored in too. For me, the assists are a first-order approximation to the overall defensive difficulty but I would not determine a positional adjustment solely on that. It's only the combination of the huge disparity in assists between 2b and 3b, together with the claim that 3b was nevertheless *more* important that cause me to question things.

Finally, there are bunts. I'm sure there were many more bunts in the deadball era. However, those are already included in the assist totals. *Even with* all those bunts, 3b were making 1000 fewer plays/year league-wide.

SS are just as responsible as 2b for turning the DP

It's somewhat close, but I doubt that has ever been true. Surely there are many more 5-4-3 DPs than 3-6-3/3-6-1.

Glove technology is also partly responsible for the decreasing number of bunts. I've argued before that contemporaries were blown away by Pie Traynor but in reality he was just the beginning of a long line of third basemen who learned how to play the position better with a modern glove. They only knew looking back, they couldn't see looking forward.

@58: I was thinking there were lots of 4-6-3 DPs but I didn't look it up. And those aren't as hard anyway, so I probably overstated it.

I also think that using offensive performance to assess defensive difficulty is conceptually wrong even if it's sometimes pragmatically necessary....

But why is it conceptually wrong? Other than your own theory on 3b, is there any other position at any time in history where the position was difficult to field but those players hit better than average? Or vice versa where it was easy to field and yet hitters were worse than average?

OK I'll try to answer that. You could say DH but thats so hard to figure because of the way its used to rotate injured/tired players etc. and because its hard to evaluate the effect of sitting on a bench. And they dont field.

Well what about LF? it should be easier to play than RF because of the throw. Different park configurations might make it harder to play LF but most parks are pretty symmetrical. They should definitely hit better than RFers. But they seem almost dead even (from 1958-78 LF is about half run better per the baseballref link). so what is happening? To me the only explanation is that at some pt you run out of front line talent. Maybe there are 25 guys are well qualified to play CF in all of baseball and another 25 or so who are well qualified to play RF and then you start to run out. There really arent 25 more guys who are fast enuf and still hit really well to play LF, they've been already used up filling up RF and CF etc.

So you maybe 10 guys are idea for LF and the rest of them are simply slightly lower level but thats all you got left so you stick them in LF. So LF dont really hit as well as they should in theory because there isnt enuf talent left over to take advantage of the easier requirements.

So that to me would be a better explanation of what is going on there is some shortage of talent in the infield, most of the best guys are already assigned to 2b SS and you somewhat lower level guys filling out the 3b ranks. So even though 3b is easier as you suggest, there isnt enuf talen available to take advantage of that ease and find better hitters.

There's also financial pressures, independent leagues, etc. during this time which probably made it harder to truly fill every position with the best talent. SO I guess its possible 3b suffers from some sort of dearth of talent. I wish there was a better way to measure it.

Finally, there are bunts. I'm sure there were many more bunts in the deadball era. However, those are already included in the assist totals. *Even with* all those bunts, 3b were making 1000 fewer plays/year league-wide.

Yeah I realize that. Its a huge number and that helps make your case.

They've only lost about 0.65 bunts per game/team from 1918 to 1930 (from 1.2 to .55). 3b is getting more of those than the other positions but I think its less than half, perhaps 40% of bunts are field by 3b as I recall. So 40 bunts a year/team are lost. Multiply by 16 teams somewhere almost 650 less bunts are fielded by 3b in the lively ball era.

Thats a large chunk of those 1000 plays per year, where are the other 350 coming from?

In 1919; fielded bunts resulting in outs: 3b 273, 1b 191, C 181

Or vice versa where it was easy to field and yet hitters were worse than average?

Yes. 1b in the deadball era. I'm not sure that they hit "worse than average", but far worse than they did by the 1920s.

So that to me would be a better explanation of what is going on there is some shortage of talent in the infield, most of the best guys are already assigned to 2b SS and you somewhat lower level guys filling out the 3b ranks. So even though 3b is easier as you suggest, there isnt enuf talen available to take advantage of that ease and find better hitters.

This is certainly possible and of course more possible when MLB talent was relatively thin (as it was in the deadball era, what with a new league and much lower population to draw from than later). But I don't know how to measure that, much less prove it.

So 40 bunts a year/team are lost. Multiply by 16 teams somewhere almost 650 less bunts are fielded by 3b in the lively ball era.

Thats a large chunk of those 1000 plays per year, where are the other 350 coming from?

My numbers were NL only (I can't remember why I did that; must have been an issue at the time), so it would be 325 fewer bunts fielded by 3b and leaving 2b making 675 more assists every year. And the only reason I can think of for that is to say that lots more balls were (and are) hit to the general area of 2b than to 3b.

bunts fielded seem to break down: P 52% 3b 21% 1b 14% C 14%. at least in that year.

Yes. 1b in the deadball era. I'm not sure that they hit "worse than average", but far worse than they did by the 1920s.

Yeah thats interesting. You dont see the Jim Bottomleys and Jacque Fourniers emerge there until the 1920s.

BUt was that because of glove technology? We've seen that mentioned up thread. One could counter argue that the reason Hal Chase was so valuable was his ability to field 1b with the crappy gloves they had at the time.

I dont think we can assume fielding 1b was as easy in the deadball era.

I believe the number of double plays turned by 2B went up as the number of bunts declined.

@65: Fielding absolutely was more difficult, at least in one sense, in the deadball era: way more errors. And the improved gloves reduced those a lot. In, say, 1903 about 30% of all runs scored came on errors. By 1930 that was down to 15%.

@66: That's absolutely correct, though I wouldn't call it a 1-1 correlation. In 1905-1909, the high in DPs for all NL 2b combined was 429 and for 3b combined it was 159. By 1926-1930 those numbers were 877 and 228. Not all of that is the decline in bunts; partly it was the fact that batters were hitting the ball harder and making it thereby easier to turn the DP assuming no errors.

was going to suggest using errors as a proxy for how hard the position is to play. That would obviously help Mephisto's thesis a good bit as 2b has considerably more errors than 3b at this time on the order of 25-30%. Well that's take a look at that:


total MLB errors at 2b vs 3b 1918-1925

year.....2b...... 3b

1918.. 298.. 238
1919.. 294.. 233
1920.. 408.. 355
1921.. 466.. 346
1922.. 428.. 330
1923.. 461.. 361
1924.. 498.. 343
1925.. 482.. 422

wow that's just a gigantic jump corresponding pretty much exactly with the introduction of the lively ball. In two years from 1919 to 1921 its a 58.5% increase in 2b errors. And probably not from any special positioning or some different role, 3b errors are up 57% in that same period. Its like rockets are now shooting through the infield.

I didnt look at NL numbers in particular but that would be interesting because I guess the lively ball wasnt deployed until a season later? is that correct?

the 1920 total errors at 3b are split nearly evenly 173 NL 182 AL, so nothing there I suppose.

My recollection is that both leagues introduced the lively ball at the same time (and just as important banned the spitter and required clean balls in play). However, the NL eliminated the lively ball before the AL did.

I feel certain I read that the AL introduced it first perhaps in late 1919? Doesnt mean its true though.

Do we have any idea how often players got bunt hits in the dead ball era?

I want to say thanks to y'all having this discussion on infield play! It's really interesting to read!

Do we have any idea how often players got bunt hits in the dead ball era?

I don't, sorry.

@68:

Don't forget that 1918 and 1919 were both shortened seasons.

Also, I believe the offensive jump was mostly due to outlawing putting sh!t on the ball. Also, keeping clean balls in play. I don't think the ball was made more lively.

Yes I've heard that too Dan but isnt this season when we first hear about ''tape measure'' hrs? Like I think Ruth hit a 515' hr in spring training that year. The ball seems to carry further

My recollection from Bill James is that the ball was livened in 1910/11, but offense had only a brief upswing because the spitball restored the advantage to pitchers. He said (all this is off memory) that the ball wasn't made livelier in 1920, but as Dan said offense increased because the spitball was banned and new balls were put in play more often. OTOH, the traditional account says the lively ball was introduced in 1920, so I don't know for sure which is right.

Don't forget that 1918 and 1919 were both shortened seasons.

I forgot that so that changes everything with respect to the errors at 3b/2b.

a look at baseballs through the years

COR throughout the years.

2019: .485
2018: .485
2017: .557
2016: .556 .494 Lichtman commissioned study in which the balls are fired at 120 mph-CCOR
2015: .558
2014: .554
2014-2015: .489 Lichtman commissioned study in which the balls are fired at 120 mph-CCOR
2013: .553
2012: .557
2011: .562
2010: .565
2009: .566
2008: .566
2007: .568
2006: .559
2005: .567
2004: .572
2003: .567
2000: .554
1999: .548
1998: .551
1982: .57
1980-1976: .53 to .56*****
1977: .563
1976: .556
1973: .559
1970: .556
1963: .559
1961: .5638 to .574 Popular Mechanics tested 12 balls as well and got a range of .50 to .68 while using a bat.
1960: .5517**
1953: .569
1952: .548**
1943: .42 for reclaimed rubber. .40 for balata*
1938: .46*
1936: .5672***
1927: .5534****
1925: .56
1923: .57
1914: .56


*: Briggs tested the balls by flinging them at 104 mph while MLB tests them at 58 mph
**: Balls tested were 1 year old.
***: 25 year old tested ball
****: 34 year old tested ball
*****: Exact year unknown since Charlie Finley's family donated them 25 years later

this has been a great discussion Mephisto and I do think you make a very good case for your position. I am going to try to find some time to look at 1bmen fielding during this period as well to see if there may be reason to think the corner positions became easier when the lively ball was introduced.

Thank you ss. I really enjoyed it.

My recollection from Bill James is that the ball was livened in 1910/11, but offense had only a brief upswing because the spitball restored the advantage to pitchers. He said (all this is off memory) that the ball wasn't made livelier in 1920, but as Dan said offense increased because the spitball was banned and new balls were put in play more often. OTOH, the traditional account says the lively ball was introduced in 1920, so I don't know for sure which is right.

The Spalding "cork center" ball was patented in August of 1909, but if you go by the adverts in the Spalding Guides, it wasn't used until the 1911 season. And the 1920 Guide has the same advert for the ball as it did the year before. Bill and Dan are right.

But while it's true that after the Chapman tragedy balls were put it play "more often", the actual turnover of baseballs was nothing like it is today, when a single pitch in the dirt causes a new ball to be put in play. I'm not exactly sure when this practice started, but it wasn't that long ago that the same ball could be used for several plays in a row where there weren't any foul balls. There wasn't the current obsession with always having a brand new ball.

And the 1920 Guide has the same advert for the ball as it did the year before.

but what exactly does that prove? They simply didnt bother to hire an artist for a new ad. That really doesnt say anything about the composition.

And what about 1913? Did they change ads then? Clearly something about that time that deadened the ball as the game resorted to the previous dead ball environment.